Struct ExplodeSpec 

pub struct ExplodeSpec {
    pub field: String,
    pub table: String,
    pub primary_key: Vec<String>,
    pub columns: Vec<ExplodeColumn>,
    pub track_history: bool,
}

v0.31 / #40: declarative “flatten this JSON array into a real SQLite table” spec on an inventory manifest. The projector creates the table on first registration (CREATE TABLE IF NOT EXISTS + indexes) and writes a row per element of payload[field] on every result, scoped by (pc_id, job_id) so each PC’s rows replace cleanly without a per-PC schema.

Fields

field: String

JSON array key under the payload to explode. E.g. "apps" for payload: { apps: [{...}, {...}] }.

table: String

Derived SQLite table name. Operators choose this — pick something namespaced + stable (inventory_sw_apps, not apps) so multiple inventory manifests don’t collide on a generic name.

primary_key: Vec<String>

Element-level fields that uniquely identify a row inside one PC’s payload. The full PK is (pc_id, job_id) + these columns. Required — operators must think about uniqueness (e.g. ["name", "source"] for installed apps because the same name appears in multiple uninstall hives).

v0.31 / #41: same tuple drives history identity. When track_history is on, the projector serialises these fields’ values into inventory_history.identity_json for every change event, so queries like “every PC that ever installed Chrome (any source)” filter on identity_json content without a per-manifest schema.

columns: Vec<ExplodeColumn>

Per-element fields that become columns in the derived table.

track_history: bool

v0.31 / #41: when true (default false), the projector diffs each PC’s incoming payload against the prior rows for the same (pc_id, job_id) BEFORE the DELETE-then-INSERT replace, and writes added / removed / changed events into inventory_history. Lets operators answer time-dimension questions (“when did Chrome 120 first appear on PC X?”, “what’s the Win 11 23H2 rollout curve”) without storing per-scan snapshots. Off by default so operators opt in per-spec — history has a real storage cost on long-lived deployments (mitigated by the 90-day default retention sweeper, see cleanup module).

Trait Implementations

impl Clone for ExplodeSpec

fn clone(&self) -> ExplodeSpec

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ExplodeSpec

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for ExplodeSpec

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl JsonSchema for ExplodeSpec

fn schema_name() -> Cow<'static, str>

The name of the generated JSON Schema.

fn schema_id() -> Cow<'static, str>

Returns a string that uniquely identifies the schema produced by this type.

fn json_schema(generator: &mut SchemaGenerator) -> Schema

Generates a JSON Schema for this type.

fn inline_schema() -> bool

Whether JSON Schemas generated for this type should be included directly in parent schemas, rather than being re-used where possible using the $ref keyword.

impl Serialize for ExplodeSpec

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.